Return-path: Received: from mail-wg0-f45.google.com ([74.125.82.45]:34956 "EHLO mail-wg0-f45.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751385AbbEUIoQ convert rfc822-to-8bit (ORCPT ); Thu, 21 May 2015 04:44:16 -0400 Received: by wgfl8 with SMTP id l8so78084501wgf.2 for ; Thu, 21 May 2015 01:44:15 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <1432197058.2343.7.camel@sipsolutions.net> References: <1431349503-5461-1-git-send-email-michal.kazior@tieto.com> <1431508609-9841-1-git-send-email-michal.kazior@tieto.com> <1431508609-9841-2-git-send-email-michal.kazior@tieto.com> <1432127653.19214.12.camel@sipsolutions.net> <1432197058.2343.7.camel@sipsolutions.net> Date: Thu, 21 May 2015 10:44:15 +0200 Message-ID: (sfid-20150521_104422_110041_8AAC7E86) Subject: Re: [PATCH v2 2/2] mac80211: prevent possible crypto tx tailroom corruption From: Michal Kazior To: Johannes Berg Cc: linux-wireless Content-Type: text/plain; charset=UTF-8 Sender: linux-wireless-owner@vger.kernel.org List-ID: On 21 May 2015 at 10:30, Johannes Berg wrote: > On Thu, 2015-05-21 at 10:16 +0200, Michal Kazior wrote: > >> >> rtnl_lock(); >> >> + list_for_each_entry(sdata, &local->interfaces, list) >> >> + cancel_delayed_work_sync(&sdata->dec_tailroom_needed_wk); >> > >> > Would it make sense to just flush the work here? That way we don't have >> > to do all the other things. >> >> Hmm.. dec_tailroom_needed_wk is queued on system workqueue now so >> there's no feasible way of flushing it (restart_work is on a system >> workqueue as well). It'd need to be moved to local->workqueue. I guess >> that would work too. > > flush_work()? Oh. I wasn't aware of this call.. Thanks for pointing out :-) Michał